Building Intel(R) MPI Benchmarks

Building Intel(R) MPI Benchmarks#

Linux* OS#

To build the benchmarks for Linux* OS, do the following:

  1. Set up the environment for the compiler and Intel(R) MPI Library.

    • For the Intel(R) compilers, run:

      source <compiler-dir>/env/vars.sh intel64

    • For the Intel(R) MPI Library, run:

      source <intel-mpidir>/env/vars.sh

  2. Set the CC variable to point to the appropriate compiler wrapper, mpiicx, or mpicx.

  3. For the GPU support:

    oneAPI Level-Zero API 1.0. CUDA Version 11.3 or higher.

    • Set CUDA_INCLUDE_DIR to the directory containing the CUDA include files.

    export CUDA_INCLUDE_DIR=<path-to-include-files>
    
    • Set ZE_INCLUDE_DIR to the directory containing the Level Zero include files:

    export ZE_INCLUDE_DIR=<path-to-include-files>
    
  4. Run one or more Makefile commands listed in the table.

Command

Description

make clean

Remove legacy binary object files and executable files.

make IMB-MPI1

Build the executable file for the IMB-MPI1 component.

make IMB-MPI1-GPU

Build the executable file for the IMB-MPI1-GPU benchmarks.

make IMB-EXT

Build the executable file for one-sided communications benchmarks.

make IMB-IO

Build the executable file for I/O benchmarks.

make IMB-NBC

Build the executable file for IMB-NBC benchmarks.

make IMB-RMA

Build the executable file for IMB-RMA benchmarks.

make all

Build all executable files available.

Windows* OS#

To build the benchmarks, use the enclosed Microsoft Visual Studio* solution files located in version-specific subdirectories under the imb/WINDOWS directory.

In Visual Studio, press F7 or go to Build > Build Solution to create an executable.

See Also

Running Intel(R) MPI Benchmarks